Record everything in one place. From your lab metadata and specific equipment configurations to trial and experimental subject records, DataJoint is your all-in-one repository.
Schedule a demoLorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique.
Process and synchronize parallel data sources, extract relevant information, and evaluate each side-by-side on the same time clock, regardless of instrument or modality.
Book a demo
Merge team member workflows without conflict, analyze data streams individually or as a whole, and set the stage for publishing and reproducing results.
Learn moreNulla vitae elit libero, a pharetra augue. Duis mollis, est non commodo luctus, nisi erat porttitor ligula, eget lacinia odio sem nec elit. Cras mattis consectetur purus sit amet fermentum. Donec id elit non mi porta gravida at eget metus. Etiam porta sem malesuada magna mollis euismod.
Dr. Hui-Chen Lu, Professor
Gill Center for Neuroscience
Dr. Lauri Nurminen, Assistant Professsor
University of Houston
Dr. Sandeep Robert Datta, Professor
Harvard Medical School
DataJoint Works is a general-purpose data operations platform. However its roots lie in supporting neuroscience studies, with extensive use in experiments that combine instruments for electrophysiology, stimuli, multiphoton microscopy of neuronal signals, optogenetics, behavior, histology, and many others. A collection of reference implementations – DataJoint Elements – supports a variety of data-rich studies.
Having an existing data pipeline is even better. There will be some development effort required to reorganize the existing pipeline into DataJoint tables to define proper data model and computational dependencies, but all existing processing/analysis code is fully reusable.
DataJoint requires python proficiency at the same level as other scientific packages in python (such as numpy, pandas, matplotlib, etc.).
Some basic knowledge of database design and operation is preferred (e.g. primary key, foreign key, joins, normalization, etc.).
Materials to learn the basics of DataJoint can be found at DataJoint Python and DataJoint Tutorials
Reference implementations of data pipelines for various neurophysiology data modalities and analyses can be found at DataJoint Elements
The large data files (e.g. raw files) and bulky processed results are stored as files on cloud object storage or on-premise file servers, with points to these files managed seamlessly within the DataJoint framework.
The metadata and/or computational results (smaller in size) are stored directly within the database (MySQL). The database server itself can be hosted on the cloud or on-premise.
Access and interaction with the data can be done via programming APIs in python (DataJoint-python) or MATLAB (DataJoint-MATLAB).
Furthermore, as DataJoint is built on top of relational databases (MySQL), users can use any SQL-supported tools to access the data.
